See2Act couples action denoising with viewpoint refinement in a diffusion-based imitation learning policy trained on keyframe-anchored camera poses, recovering informative views under occlusion and improving RLBench performance by up to 34% with zero-shot sim-to-real transfer.
hub Canonical reference
Robotic Control via Embodied Chain-of-Thought Reasoning
Canonical reference. 100% of citing Pith papers cite this work as background.
59
Pith papers citing it
Background
100% of classified citations
abstract
A key limitation of learned robot control policies is their inability to generalize outside their training data. Recent works on vision-language-action models (VLAs) have shown that the use of large, internet pre-trained vision-language models as the backbone of learned robot policies can substantially improve their robustness and generalization ability. Yet, one of the most exciting capabilities of large vision-language models in other domains is their ability to reason iteratively through complex problems. Can that same capability be brought into robotics to allow policies to improve performance by reasoning about a given task before acting? Naive use of "chain-of-thought" (CoT) style prompting is significantly less effective with standard VLAs because of the relatively simple training examples that are available to them. Additionally, purely semantic reasoning about sub-tasks, as is common in regular CoT, is insufficient for robot policies that need to ground their reasoning in sensory observations and the robot state. To this end, we introduce Embodied Chain-of-Thought Reasoning (ECoT) for VLAs, in which we train VLAs to perform multiple steps of reasoning about plans, sub-tasks, motions, and visually grounded features like object bounding boxes and end effector positions, before predicting the robot action. We design a scalable pipeline for generating synthetic training data for ECoT on large robot datasets. We demonstrate, that ECoT increases the absolute success rate of OpenVLA, the current strongest open-source VLA policy, by 28% across challenging generalization tasks, without any additional robot training data. Additionally, ECoT makes it easier for humans to interpret a policy's failures and correct its behavior using natural language.
hub tools
citation-role summary
background 13
citation-polarity summary
roles
background 13polarities
background 13representative citing papers
PearlVLA achieves SOTA on LIBERO by separating VLM representations into visual grounding and an iterative latent plan branch refined via world model queries and RefineNet with process-reward RL.
PhAIL provides an open benchmark and distributional evaluation method for real-robot VLA policies using time-to-success CDF, HRT scoring, and KS significance tests.
AwareVLN introduces a structural reasoning module and automatic data engine with progress division to equip VLN agents with self-awareness of agent state and task progress, outperforming prior methods on Habitat datasets.
CrossVLA develops a surrogate log-probability estimator for DPO on flow-matching VLAs, shows DoRA outperforming LoRA by +10.4 pp mean on LIBERO, and identifies inference bottlenecks with limited caching gains.
CoRAL lets LLMs act as adaptive cost designers for motion planners while using VLM priors and online identification to handle unknown physics, achieving over 50% higher success rates than baselines in unseen contact-rich robotic scenarios.
A multimodal transformer generates and caches interleaved text-image traces to guide closed-loop actions, achieving 92.4% success on LIBERO-Long and 95.5% average on LIBERO.
Being-H0.7 adds future-aware latent reasoning to direct VLA policies via dual-branch alignment on latent queries, matching world-model benefits at VLA efficiency.
π₀.₇ is a steerable generalist robotic model that uses rich multimodal prompts including language, subgoal images, and performance metadata to achieve out-of-the-box generalization across tasks and robot bodies.
E-TTS introduces a plug-and-play test-time scaling method for embodied tasks that unifies reasoning-action sampling with history buffers and closed-loop refinement to improve performance on manipulation benchmarks.
Vesta is a unified embodied generalist model that outperforms specialist baselines by over 20% on average and improves real-world robotic task success by over 35%.
Qwen-RobotManip applies unified alignment across representation, motion, and behavior to enable large-scale training on heterogeneous manipulation data, yielding emergent generalization on out-of-distribution robotic benchmarks.
DIRECT is a multimodal-context router that allocates test-time compute across chain-of-thought depth, model size, and memory history for VLM embodied planners, improving the success-cost Pareto frontier and matching stronger models at up to 65% lower latency on benchmarks and a physical Franka arm.
ERVLA trains on a 978k-trajectory embodied CoT corpus using reasoning as supervision with dropout, then predicts actions without CoT at test time, reaching 86.9% on LIBERO-Plus and 53.2% on VLABench.
VISUALTHINK-VLA uses visual evidence tokens and selective routing to reach top success rates on VLA benchmarks while cutting reasoning latency from multi-second to sub-second levels.
DeMiAn re-annotates robot and egocentric videos with VLM-generated dense labels across motion, scene, pose, and reasoning aspects, then uses a learned instructor to boost policy success by 5 points on RoboCasa over task-only baselines.
UAM adds a Dorsal Expert initialized from a generative model and trained on visual dynamics prediction to preserve over 95% of VLM multimodal ability in VLA training while achieving top success rates on manipulation tasks including OOD cases.
MolmoAct2 is an open VLA model that outperforms baselines like Pi-05 on 7 benchmarks and whose backbone surpasses GPT-5 on 13 embodied-reasoning tasks through new datasets, specialized training, and architecture changes for lower latency.
VLA-ATTC equips VLA models with adaptive test-time compute via an uncertainty clutch and relative action critic, cutting failure rates by over 50% on LIBERO-LONG.
Sentinel-VLA adds metacognitive status monitoring to VLA models for on-demand reasoning and error recovery, reporting over 30% higher real-world task success than prior SOTA.
LaST-R1 introduces a RL post-training method called LAPO that optimizes latent Chain-of-Thought reasoning in vision-language-action models, yielding 99.9% success on LIBERO and up to 22.5% real-world gains.
GazeVLA pretrains on large human egocentric datasets to capture gaze-based intention, then finetunes on limited robot data with chain-of-thought reasoning to achieve better robotic manipulation performance than baselines.
Denoising Recursion Models train multi-step noise reversal in looped transformers and outperform the prior Tiny Recursion Model on ARC-AGI.
Sim-and-real co-training for robot policies is driven primarily by balanced cross-domain representation alignment and secondarily by domain-dependent action reweighting.
citing papers explorer
-
Learning to See While Learning to Act: Diffusion Models for Active Perception in Robot Imitation
See2Act couples action denoising with viewpoint refinement in a diffusion-based imitation learning policy trained on keyframe-anchored camera poses, recovering informative views under occlusion and improving RLBench performance by up to 34% with zero-shot sim-to-real transfer.
-
PearlVLA: Progressive Embodied Action-Plan Refinement in Latent Space
PearlVLA achieves SOTA on LIBERO by separating VLM representations into visual grounding and an iterative latent plan branch refined via world model queries and RefineNet with process-reward RL.
-
PhAIL: A Real-Robot VLA Benchmark and Distributional Methodology
PhAIL provides an open benchmark and distributional evaluation method for real-robot VLA policies using time-to-success CDF, HRT scoring, and KS significance tests.
-
AwareVLN: Reasoning with Self-awareness for Vision-Language Navigation
AwareVLN introduces a structural reasoning module and automatic data engine with progress division to equip VLN agents with self-awareness of agent state and task progress, outperforming prior methods on Habitat datasets.
-
CoRAL: Contact-Rich Adaptive LLM-based Control for Robotic Manipulation
CoRAL lets LLMs act as adaptive cost designers for motion planners while using VLM priors and online identification to handle unknown physics, achieving over 50% higher success rates than baselines in unseen contact-rich robotic scenarios.
-
Being-H0.7: A Latent World-Action Model from Egocentric Videos
Being-H0.7 adds future-aware latent reasoning to direct VLA policies via dual-branch alignment on latent queries, matching world-model benefits at VLA efficiency.
-
E-TTS: A New Embodied Test-Time Scaling Framework for Robotic Manipulation
E-TTS introduces a plug-and-play test-time scaling method for embodied tasks that unifies reasoning-action sampling with history buffers and closed-loop refinement to improve performance on manipulation benchmarks.
-
Vesta: A Generalist Embodied Reasoning Model
Vesta is a unified embodied generalist model that outperforms specialist baselines by over 20% on average and improves real-world robotic task success by over 35%.
-
Qwen-RobotManip Technical Report: Alignment Unlocks Scale for Robotic Manipulation Foundation Models
Qwen-RobotManip applies unified alignment across representation, motion, and behavior to enable large-scale training on heterogeneous manipulation data, yielding emergent generalization on out-of-distribution robotic benchmarks.
-
DIRECT: When and Where Should You Allocate Test-Time Compute in Embodied Planners?
DIRECT is a multimodal-context router that allocates test-time compute across chain-of-thought depth, model size, and memory history for VLM embodied planners, improving the success-cost Pareto frontier and matching stronger models at up to 65% lower latency on benchmarks and a physical Franka arm.
-
Revisiting Embodied Chain-of-Thought for Generalizable Robot Manipulation
ERVLA trains on a 978k-trajectory embodied CoT corpus using reasoning as supervision with dropout, then predicts actions without CoT at test time, reaching 86.9% on LIBERO-Plus and 53.2% on VLABench.
-
How to Instruct Your Robot: Dense Language Annotations Power Robot Policy Learning
DeMiAn re-annotates robot and egocentric videos with VLM-generated dense labels across motion, scene, pose, and reasoning aspects, then uses a learned instructor to boost policy success by 5 points on RoboCasa over task-only baselines.
-
MolmoAct2: Action Reasoning Models for Real-world Deployment
MolmoAct2 is an open VLA model that outperforms baselines like Pi-05 on 7 benchmarks and whose backbone surpasses GPT-5 on 13 embodied-reasoning tasks through new datasets, specialized training, and architecture changes for lower latency.
-
VLA-ATTC: Adaptive Test-Time Compute for VLA Models with Relative Action Critic Model
VLA-ATTC equips VLA models with adaptive test-time compute via an uncertainty clutch and relative action critic, cutting failure rates by over 50% on LIBERO-LONG.
-
Sentinel-VLA: A Metacognitive VLA Model with Active Status Monitoring for Dynamic Reasoning and Error Recovery
Sentinel-VLA adds metacognitive status monitoring to VLA models for on-demand reasoning and error recovery, reporting over 30% higher real-world task success than prior SOTA.
-
LaST-R1: Reinforcing Robotic Manipulation via Adaptive Physical Latent Reasoning
LaST-R1 introduces a RL post-training method called LAPO that optimizes latent Chain-of-Thought reasoning in vision-language-action models, yielding 99.9% success on LIBERO and up to 22.5% real-world gains.
-
GazeVLA: Learning Human Intention for Robotic Manipulation
GazeVLA pretrains on large human egocentric datasets to capture gaze-based intention, then finetunes on limited robot data with chain-of-thought reasoning to achieve better robotic manipulation performance than baselines.
-
A Mechanistic Analysis of Sim-and-Real Co-Training in Generative Robot Policies
Sim-and-real co-training for robot policies is driven primarily by balanced cross-domain representation alignment and secondarily by domain-dependent action reweighting.
-
A1: A Fully Transparent Open-Source, Adaptive and Efficient Truncated Vision-Language-Action Model
A1 is a transparent VLA framework achieving state-of-the-art robot manipulation success with up to 72% lower latency via adaptive layer truncation and inter-layer flow matching.
-
ThermoAct:Thermal-Aware Vision-Language-Action Models for Robotic Perception and Decision-Making
ThermoAct integrates thermal imaging into VLA models via a VLM planner to enable robots to perceive physical properties like heat and improve safety over vision-only systems.
-
PALM: Progress-Aware Policy Learning via Affordance Reasoning for Long-Horizon Robotic Manipulation
PALM improves long-horizon robotic manipulation success by distilling affordance representations for object interaction and predicting within-subtask progress in a VLA model.
-
Clutter-Robust Vision-Language-Action Models through Object-Centric and Geometry Grounding
OBEYED-VLA improves VLA robustness in cluttered real-world manipulation by disentangling perception into VLM-based object-centric grounding and geometry-aware stages, then fine-tuning the policy only on single-object demonstrations.
-
mimic-video: Video-Action Models for Generalizable Robot Control Beyond VLAs
mimic-video combines internet video pretraining with a flow-matching decoder to achieve state-of-the-art robotic manipulation performance with 10x better sample efficiency than vision-language-action models.
-
InternVLA-M1: A Spatially Guided Vision-Language-Action Framework for Generalist Robot Policy
InternVLA-M1 uses spatially guided pre-training on 2.3M examples followed by action post-training to deliver up to 17% gains on robot manipulation benchmarks and 20.6% on unseen objects.
-
SimpleVLA-RL: Scaling VLA Training via Reinforcement Learning
SimpleVLA-RL applies tailored reinforcement learning to VLA models, reaching SoTA on LIBERO, outperforming π₀ on RoboTwin, and surpassing SFT in real-world tasks while reducing data needs and identifying a 'pushcut' phenomenon.
-
Policy Contrastive Decoding for Robotic Foundation Models
PCD redirects robotic policies toward object-relevant visual features via contrastive decoding on masked inputs, improving generalization without retraining or weight access.
-
GraspVLA: a Grasping Foundation Model Pre-trained on Billion-scale Synthetic Action Data
GraspVLA shows that pretraining a grasping model on a billion synthetic action frames enables zero-shot open-vocabulary performance and sim-to-real transfer.
-
Hi Robot: Open-Ended Instruction Following with Hierarchical Vision-Language-Action Models
A hierarchical VLA architecture lets robots follow complex instructions and situated feedback by separating high-level reasoning from low-level control.
-
A Survey on Vision-Language-Action Models for Embodied AI
This is the first survey on vision-language-action models, providing a taxonomy across three lines, plus summaries of datasets, simulators, benchmarks, challenges, and future directions in embodied AI.
-
Position: Vision-Language-Action Models Cannot Be Verified to Perform Physical Reasoning
VLA benchmark success rates cannot distinguish semantic generalization from physical reasoning due to an identifiability gap in current evaluation protocols.
-
Learning Action Priors for Cross-embodiment Robot Manipulation
A two-stage framework pretrains an action module with temporal motion priors from unconditioned trajectories using flow-matching, then transfers it to VLA training via decoder reuse and distillation, yielding better performance on cross-embodiment tasks.
-
LaST-HD: Learning Latent Physical Reasoning from Scalable Human Data for Robot Manipulation
LaST-HD creates a shared latent dynamics space via a world model to transfer physical reasoning from scalable human-hand demonstrations to robots, achieving over 90% accuracy with 20 minutes of new data after mixed training.
-
Coarse-to-Control: Action-Token Planning for Vision-Language-Action Models
Coarse-to-Control adds planning via coarse action tokens in the same vocabulary as control actions, improving VLA performance on long-horizon manipulation tasks.
-
Dreaming when Necessary: Advancing World Action Models with Adaptive Multi-Modal Reasoning
AdaWAM introduces an adaptive router that triggers textual or visual reasoning as needed in world action models, claiming better efficiency and performance than prior embodied policies on simulated and real tasks.
-
World-Language-Action Model for Unified World Modeling, Language Reasoning, and Action Synthesis
WLA models use an autoregressive Transformer to jointly predict textual subtasks, subgoal images, and robot actions from instructions, images, and states, reporting SOTA success rates on RoboTwin2.0 and RMBench.
-
Wall-OSS-0.5 Technical Report
Wall-OSS-0.5 is a 4B VLA model pretrained across many embodiments that achieves zero-shot real-robot performance on a 17-task suite and outperforms π_0.5 after fine-tuning.
-
VLA-GSE: Boosting Parameter-Efficient Fine-Tuning in VLA with Generalized and Specialized Experts
VLA-GSE uses spectral decomposition of the VLA backbone to create generalized and specialized experts, enabling effective robot task adaptation while updating only 2.51% of parameters and achieving 81.2% zero-shot success on LIBERO-Plus.
-
ReFineVLA: Multimodal Reasoning-Aware Generalist Robotic Policies via Teacher-Guided Fine-Tuning
ReFineVLA adds teacher-generated reasoning steps to VLA training and reports state-of-the-art success rates on SimplerEnv WidowX and Google Robot benchmarks.
-
CoEnv: Driving Embodied Multi-Agent Collaboration via Compositional Environment
CoEnv introduces a compositional environment that integrates real and simulated spaces for multi-agent robotic collaboration, using real-to-sim reconstruction, VLM action synthesis, and validated sim-to-real transfer to achieve high success rates on multi-arm manipulation tasks.
-
Hierarchical DLO Routing with Reinforcement Learning and In-Context Vision-language Models
Hierarchical framework pairs in-context VLMs for high-level plan synthesis with RL-trained low-level skills and failure recovery to reach 92% success on long-horizon DLO routing across varied scenes and language inputs.
-
A Careful Examination of Large Behavior Models for Multitask Dexterous Manipulation
Multi-task pretraining of diffusion policies on diverse robot data produces more successful, robust, and data-efficient policies for dexterous manipulation than single-task baselines, with performance scaling with pretraining size and diversity.
-
NORA: A Small Open-Sourced Generalist Vision Language Action Model for Embodied Tasks
NORA is a compact 3B-parameter VLA model trained on 970k robot demonstrations that outperforms larger VLA models in embodied tasks while using significantly less computational resources.
-
SafeVLA: Towards Safety Alignment of Vision-Language-Action Model via Constrained Learning
SafeVLA applies constrained reinforcement learning via CMDP min-max optimization to VLAs, cutting safety violation costs by 83.58% while preserving task success on long-horizon mobile manipulation tasks.
-
What Matters in Building Vision-Language-Action Models for Generalist Robots
Systematic tests of VLM backbones, policy architectures, and cross-embodiment data yield RoboVLMs that set new SOTA on robot manipulation benchmarks while requiring few manual designs.
-
Hierarchical Prompting with Dual LLM Modules for Robotic Task and Motion Planning
A dual-LLM hierarchical framework for robotic task and motion planning, integrating object detection, achieves 86% success across 24 test scenarios ranging from simple spatial commands to infeasible requests.
- CodeGraphVLP: Code-as-Planner Meets Semantic-Graph State for Non-Markovian Vision-Language-Action Models